Real-time whole slide pathology image cell counting

Abstract

Techniques are provided for determining a cell count within a whole slide pathology image. The image is segmented using a global threshold value to define a tissue area. A plurality of patches comprising the tissue area are selected. Stain intensity vectors are determined within the plurality of patches to generate a stain intensity image. The stain intensity image is iteratively segmented to generate a cell mask using a local threshold value that is and gradually reduced after each iteration. A chamfer distance transform is applied to the cell mask to generate a distance map. Cell seeds are determined on the distance map. Cell segments are determined using a watershed transformation, and a whole cell count is calculated for the plurality of patches based on the cell segments. A client device may be configured for real-time cell counting based on the whole cell count.

1 - 28 . (canceled) 
     
     
         29 . A real-time interactive image analysis computer-based system comprising:
 at least one computer-readable non-transitory memory storing software instructions; and   at least one processor coupled with the at least one memory and that performs the following operations upon execution of the software instructions:
 receiving and storing a digital image in the at least one memory; 
 segmenting the digital image into a plurality of patches; 
 calculating an object density value for each patch of the plurality of patches; 
 receiving a user-selected location within the digital image 
 determining patches adjacent to the user-selected location; 
 computing, in real-time, an interpolated object density value at the user-selected location based on the object density values of the patches adjacent to the user-selected location; and 
 displaying on a display the interpolated object density value at the user selected location. 
   
     
     
         30 . The system of  claim 29 , wherein the operation of computing the interpolated object density value comprises performing bilinear interpolation using the object density values of the patches adjacent to the user-selected location. 
     
     
         31 . The system of  claim 29 , wherein the patches adjacent to the user-selected location comprise at least four patches proximate to a pointer location. 
     
     
         32 . The system of  claim 29 , wherein the operation of segmenting the image comprises applying a global threshold filter to the image to identify features satisfying a threshold value. 
     
     
         33 . The system of  claim 32 , wherein the global threshold filter uses a minimum between an Otsu filter value and a Yen filter value. 
     
     
         34 . The system of  claim 29 , wherein the operations further comprise:
 receiving a selection of a region of interest within the digital image and that includes the user-selected location;   determining a boundary of the region of interest; and   calculating a total object density value for the region of interest.   
     
     
         35 . The system of  claim 34 , wherein the operation of calculating the total object density value comprises determining a percentage coverage of patches intersecting the boundary. 
     
     
         36 . The system of  claim 29 , wherein calculating the object density value comprises:
 converting each patch of the plurality of patches from an RGB space to an optical object density space;   removing pixels below a threshold intensity; and   generating intensity vectors.   
     
     
         37 . The system of  claim 29 , wherein the image comprises a whole-slide image. 
     
     
         38 . The system of  claim 37 , wherein the operation of calculating the object density value for each patch of the plurality of patches comprises counting cells within each patch of the whole slide image. 
     
     
         39 . The system of  claim 38 , wherein the operation of counting cells comprises:
 generating a cell mask by iteratively applying a threshold filter;   applying a distance transform to the cell mask;   determining cell seeds; and   performing a watershed transformation.   
     
     
         40 . The system of  claim 29 , wherein the operations further comprise:
 receiving an updated user-selected location; and   updating the displayed interpolated object density value in real-time based on the updated user-selected location.   
     
     
         41 . The system of  claim 29 , wherein the plurality of patches are randomly selected from the digital image. 
     
     
         42 . The system of  claim 29 , wherein the operation of calculating the object density value for each patch of the plurality of patches comprises applying a smoothing filter to normalize the object density values. 
     
     
         43 . The system of  claim 29 , wherein the digital image is obtained from a client device. 
     
     
         44 . The system of  claim 43 , wherein the client device comprises at least one of a tablet, cellular phone, a vehicle, or a robot. 
     
     
         45 . The system of  claim 29 , wherein the at least one processor comprises at least one of a Central Processing Unit (CPU) or a General Processing Unit (GPU). 
     
     
         46 . The system of  claim 29 , wherein the operations further comprise storing the calculated object density values in a persistent storage device that includes the at least one memory. 
     
     
         47 . The system of  claim 29 , wherein the operation of displaying comprises overlaying the interpolated density value on the digital image. 
     
     
         48 . The system of  claim 29 , wherein the at least one processor and the at least one memory are part of a distributed computer system. 
     
     
         49 . A method comprising:
 receiving and storing a digital image in the at least one memory;   segmenting the digital image into a plurality of patches;   calculating an object density value for each patch of the plurality of patches;   receiving a user-selected location within the digital image   determining patches adjacent to the user-selected location;   computing, in real-time, an interpolated object density value at the user-selected location based on the object density values of the patches adjacent to the user-selected location; and   displaying on a display the interpolated object density value at the user selected location.   
     
     
         50 . A non-transitory computer-readable medium having computer instructions stored thereon, which, when executed by at least one processor, cause the at least one processor to perform the operations of:
 receiving and storing a digital image in the at least one memory;   segmenting the digital image into a plurality of patches;   calculating an object density value for each patch of the plurality of patches;   receiving a user-selected location within the digital image   determining patches adjacent to the user-selected location;   computing, in real-time, an interpolated object density value at the user-selected location based on the object density values of the patches adjacent to the user-selected location; and   displaying on a display the interpolated object density value at the user selected location.